Applied Digital Signal Processing Exam for Electronic Engineering Students, Exams of Digital Signal Processing

A past exam paper from the cork institute of technology for the bachelor of engineering (honours) in electronic engineering program. The exam focuses on applied digital signal processing and includes questions on topics such as spectral analysis, impulse response, and adaptive equalization. Students are required to answer any four questions within two hours, with each question worth a maximum of 25 marks. The examiners are dr. Joseph connell, prof. Gerard hurley, and dr. Sean foley.

Typology: Exams

2012/2013

Uploaded on 03/30/2013

lalitesh
lalitesh 🇮🇳

4.4

(19)

108 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Cork Institute of Technology
Bachelor of Engineering (Honours) in Electronic Engineering – Award
(NFQ Level – 8)
Summer 2006
APPLIED DIGITAL SIGNAL PROCESSING
(Time: 2 Hours)
Answer any four questions
[each 25 marks]
Maximum available mark is 100.
Examiners: Dr. Joseph Connell
Prof. Gerard Hurley
Dr. Sean Foley
Q1. (a) A dual tone signal comprising a unity-amplitude, 10Hz sine wave and a unity-
amplitude, 12Hz sine wave is sampled at 100Hz. Discuss the process of using
the samples thus produced to spectrally analyse the signal accurately. [10 marks]
(b) A 16Hz, unity amplitude sinewave is sampled at 80Hz. Using a Goertzel cell
and one period of samples, evaluate the DFT for the signal in at the frequency
of interest. [15 marks]
Q2. (a) Spectral analysis of a noisy signal is required using the Welch Method. N
samples are available. Discuss the strategy involved and comment on its
performance in comparison to a straight DFT on the N samples. [10 marks]
(b) The causal impulse response of an LTI system is given as
=
)(nh {-3 1 4}.
The causal input is given as
=
)(nx {2 -1}. Show that )(*)()( lhlrlr xxxy
=
. [15 marks]
Q3. (a) File1 contains digital data sampled at s
f Hz. Describe in detail using
amplitude plots the spectral implications of creating a new file2 by inserting 2
zeros between every sample in file1.
Hence, given a file of data sampled at s
f, outline a strategy for creating a new
file of data consistent with the original signal being sampled at s
If where I is
an integer. [15 marks]
(b) A signal is sampled at 8kHz. It is required to generate a second sequence of
samples corresponding to a sampling frequency 24kHz. Write efficient
software to perform the task. Assume any filter used is 4th order. [10 marks]
pf2

Partial preview of the text

Download Applied Digital Signal Processing Exam for Electronic Engineering Students and more Exams Digital Signal Processing in PDF only on Docsity!

Cork Institute of Technology

Bachelor of Engineering (Honours) in Electronic Engineering – Award

(NFQ Level – 8)

Summer 2006

APPLIED DIGITAL SIGNAL PROCESSING

(Time: 2 Hours)

Answer any four questions [each 25 marks] Maximum available mark is 100.

Examiners: Dr. Joseph Connell Prof. Gerard Hurley Dr. Sean Foley

Q1. (a) A dual tone signal comprising a unity-amplitude, 10Hz sine wave and a unity- amplitude, 12Hz sine wave is sampled at 100Hz. Discuss the process of using the samples thus produced to spectrally analyse the signal accurately. [10 marks]

(b) A 16Hz, unity amplitude sinewave is sampled at 80Hz. Using a Goertzel cell and one period of samples, evaluate the DFT for the signal in at the frequency of interest. [15 marks]

Q2. (a) Spectral analysis of a noisy signal is required using the Welch Method. N samples are available. Discuss the strategy involved and comment on its performance in comparison to a straight DFT on the N samples. [10 marks]

(b) The causal impulse response of an LTI system is given as (^) h ( n )= {-3 1 4}. The causal input is given as (^) x ( n )={2 -1}. Show that (^) r (^) xy ( l )= rxx ( l )* h (− l ). (^) [15 marks]

Q3. (a) File1 contains digital data sampled at f (^) s Hz. Describe in detail using

amplitude plots the spectral implications of creating a new file2 by inserting 2 zeros between every sample in file1. Hence, given a file of data sampled at f (^) s , outline a strategy for creating a new file of data consistent with the original signal being sampled at If (^) s where I is an integer. (^) [15 marks]

(b) A signal is sampled at 8kHz. It is required to generate a second sequence of samples corresponding to a sampling frequency 24kHz. Write efficient software to perform the task. Assume any filter used is 4th^ order. [10 marks]

Q4. (a) Draw the block diagram for an adaptive equaliser or inverse modeller. Explain how it works. [7 marks]

(b) A first order adaptive linear combiner has input 7

x cos 2 k k

= −^ π and a desired

input (^)  

4 sin

k d (^) k

π . Calculate the optimum tapweight values and the

minimum MSE value. [18 marks]

Q5. (a) Discuss the nature of weight convergence when adapted using the LMS algorithm. In particular, include reference to the effect of the convergence

factor, μ. [7 marks]

(b) Assuming the weights in 4(b) are adapted using the LMS algorithm, calculate the weight values for the first 5 iterations starting at k = 0. Assume the initial weight values are zero and the convergence factor value is μ = 0. 1. [18 marks]